Melting of Eutectic Mixtures in Silica and Carbon Nanopores

Sep 15, 2015 - ... North Carolina State University, 911 Partners' Way, Raleigh, North ... considered to be “strongly attractive” (we expect an ele...
0 downloads 0 Views 3MB Size
Article pubs.acs.org/jced

Melting of Eutectic Mixtures in Silica and Carbon Nanopores Małgorzata Sliwinska-Bartkowiak,*,† Monika Jazdzewska,† Marcin Trafas,† Milena Kaczmarek-Klinowska,†,‡ and Keith E. Gubbins§ †

Faculty of Physics, and ‡Institute of Acoustics, Faculty of Physics, Adam Mickiewicz University, Umultowska 85, 61-614 Poznan, Poland § Department of Chemical and Biomolecular Engineering, North Carolina State University, 911 Partners’ Way, Raleigh, North Carolina 27695-7905, United States ABSTRACT: We report experimental results for the melting of eutectic mixtures confined in nanoporous matrices. Dielectric relaxation spectroscopy and differential scanning calorimetry were used to determine the solid/liquid phase diagram of C6H4Br2/CCl4 mixtures confined in controlled pore glasses (CPG) with an average pore diameter of 7.5 nm, and C6H5Br/CCl4 mixtures confined in activated carbon fibers (ACF) with a mean pore diameter of 1.4 nm. We find that the phase diagram of the confined mixtures are of the same type as that for the bulk mixture (eutectic). However, in the case of mixtures in carbon nanopores the solid/liquid coexistence lines are located at higher temperatures than for the bulk mixtures, whereas they are at lower temperatures than the bulk for mixtures confined in silica pores. These results are compared with those previously obtained for azeotropic mixtures in ACF. The results suggest that the melting temperatures, Tmp, of confined mixtures decrease relative to the bulk when the fluid−wall interactions are weaker than the fluid−fluid interaction (silica glasses), and increase in the case of fluid−wall interactions that are stronger than the fluid−fluid interaction (nanoporous carbons).



INTRODUCTION αw =

Fluid or solid phases of nanoscale dimensions, when in contact with a solid surface or confined within a narrow pore, have thermodynamic properties that are often very different from those of the macroscopic bulk material. These confinement effects arise because of reduced dimensionality and strong interaction of the molecules in the nanophase with the confining walls. For several years special attention has been focused on the freezing and melting behavior of simple fluids confined in nanoporous materials, especially in mesoporous silicas.1 It is known that for pore widths much larger than the diameter of the adsorbate molecules the shift in the melting points ΔT = Tmbulk − Tmpore can be connected with the pore width H using the Gibbs−Thomson equation.2 Many experiments performed for pores larger than 6−7 nm showed a linear relation between the shift in the melting temperature and the inverse pore width, while for smaller pore diameters deviations from the Gibbs−Thomson equation have been observed.2−5 Recently a statistical mechanical analysis of such a nanophase (the adsorbate a, molecular diameter σ) was reported5 confined within a porous material. From a corresponding states analysis of the partition function for pores of simple geometry (e.g., slitor cylinder-shaped, with pore width H) it was shown that the principal system variables are the microscopic wetting parameter, αw, and the dimensionless pore width H* = H/σ. The wetting parameter is a measure of the relative strength of the adsorbate−wall and adsorbate−adsorbate interactions, and its definition emerges from the statistical mechanical analysis as © 2015 American Chemical Society

ρε σ 2Δ s as as εaa

(1)

where ρs is the solid density in atoms per unit volume, σas, εas, and εaa are the usual size and energy parameters in the equation for the intermolecular potential energy, here taken to be the Lennard-Jones (12,6) model, and Δ is the interlayer spacing between layers of solid atoms in the substrate. The value of αw depends strongly on the atomic density of the solid, as well as on the ratio of the two interaction well depths. This wetting parameter is closely correlated with the experimental contact angles for a wide range of liquids on various substrates.6 The melting and freezing behavior of fluids confined in pores of simple geometry can be understood in terms of two main system parameters, namely the pore width H and the wetting parameter αw. It was shown4 that adsorbents with αw > 1 can be considered to be “strongly attractive” (we expect an elevation in melting temperature relatively to the bulk melting point) while those with αw < 1 are “weakly attractive” (a depression in melting temperature relatively to the bulk is expected). Very few studies of the melting of confined mixtures have been reported. Among these have been studies of confined molten salts7,8 and colloidal mixtures.9 Recently, molecular simulation and experimental studies of the confinement effect Received: February 10, 2015 Accepted: September 4, 2015 Published: September 15, 2015 3093

DOI: 10.1021/acs.jced.5b00131 J. Chem. Eng. Data 2015, 60, 3093−3100

Journal of Chemical & Engineering Data

Article

frequencies ω using a Solartron 1260 gain impedance analyzer, in the frequency range 1 Hz to 1 MHz. The sample was introduced between the capacitor plates as a suspension of nanoporous particles filled with the mixture, so that the permittivity measurements have contributions from both the bulk and confined mixtures. For dipolar liquids the melting of a solid phase is accompanied by a rapid increase of the real part of the permittivity, due to the increase of the dipolar polarization in the liquid state.11 The DSC method was applied to determine the phase transition temperatures of the confined systems by measuring the heat released in the melting process. The instrument used was a DSC Q2000 calorimeter (TA Instruments in Departmental Laboratory of Structural Research in Faculty of Physics). Temperature scanning rates of 5 K/min to 10 K/min in the range between 110 K and 330 K were used in the experiments.

on melting of azeotropic mixtures in carbon nanopores and silica glasses was reported.10,11 The confined mixtures were found to have the same type of phase diagram as the bulk mixture,12 but the melting temperature and characteristic concentrations were shifted relative to their bulk values. In this paper, we present experimental results for eutectic mixtures in both silica and carbon nanopores. The solid−liquid phase diagram of the binary mixtures is presented for both the bulk and the confined systems to investigate the influence of confinement on both the melting temperature and the character of the solid−liquid phase diagram.



EXPERIMENTAL TECHNIQUES We report experimental studies of the melting transition of two kinds of mixtures: C6H4Br2/CCl4 mixtures in controlled pores glasses (CPG), and C6H5Br/CCl4 mixtures confined in activated carbon fibers (ACF). In our experiments we have used controlled pore glasses (CPG) (made by Corning Co., New Jersey) with an average pore diameter of 7.5 nm and a pore volume of 47 cc/g. The transmission electron microscopy (TEM) image of the CPG studied is presented in Figure 1a.



EXPERIMENTAL RESULTS The melting of mixtures of C6H4Br2(2)/CCl4(1) confined in CPG, and also of C6H5Br(2)/CCl4(1) confined in ACF, as well as in the bulk phase, was determined by monitoring temperature changes in the dielectric permittivity, ε, that occur at the transition from the solid to liquid phases. The melting of dipolar liquids such as dibromobenzene is accompanied by a rapid increase in ε′ (the real part of the dielectric permittivity). Below the melting temperature, the rotational phase ceases to exist, and ε′ is almost equal to n2 (n is the refractive index) and it is only related to the deformation polarization; for the liquid phase a maximum value, ε′m, of the permittivity is observed due to the dipolar polarization of the liquid. By measuring the functions ε′(T) or C(T) (C is the electric capacity of the system) we can observe the values of ε′m and ε′s (permittivity in the solid state) in the vicinity of the melting points, and thus determine the values of the melting temperatures of the systems. From these measurements we construct the phase diagram of the system on the (T,x2)−p plane (x2 is the mole fraction of the dipolar liquid in the mixture). From the dielectric data we can also determine the relaxation processes related to molecular motion in particular phases. The electric capacity C of C6H4Br2/CCl4 mixtures was measured in the whole concentration range and over a wide range of temperatures during the melting process. Figure 2a,b,c shows C versus temperature for C6H4Br2/CCl4 bulk mixtures of various molar compositions. The transition from the solid state (characterized by the minimum values of C, which is almost temperature-independent) to the liquid state is observed over a wide temperature range, which changes with increasing mole fraction of m-dibromobenzene. For concentration x2 ≈ 0.2 a sharp increase of C(T), typical of pure liquids is observed. For x2 ≥ 0.2, the temperature range of the phase transitions regions is found to broaden. Assuming that the melting process starts at temperature T2, the temperature of the change in the monotonic character of C(T), and taking T1 as the temperature

Figure 1. (a) TEM image of CPG silica glass, (b) SEM image of ACF.

The activated carbon fibers (ACF) (obtained from Prof. K. Kaneko Laboratory, Chiba University, Japan) have roughly slitshaped pores with a mean pore width of 1.4 nm, and are built of nanocrystallites made up of defective graphene sheets (scanning electron microscopy (SEM) image in Figure 1b). Before carrying out the experiments the porous materials were heated to about 400 K and kept under vacuum (10−3 Tr) for 4 days prior to the introduction of the liquids. Carbon tetrachloride, m-dibromobenzene and bromobenzene were twice distilled; after distillation the density of the liquids was measured using a DMA 38 density meter at 298 K temperature and at normal pressure. The characterization of the liquids used is presented in Table 1. The melting of mixtures confined in CPG and ACF was determined using dielectric relaxation spectroscopy (DRS) and differential scanning calorimetry (DSC) methods.13 Dielectric measurements were performed using a parallel plate capacitor.13 The capacitance C and the tangent loss of the samples were measured at different temperatures T and

Table 1. Chemical Sample Table. Density ρ at 298 K, and Pressure p = 0.1 MPa chemical name

source

mole fraction purity

purification method

density [g/cm3]

analysis method

carbon tetrachloride m-dibromobenzene bromobenzene

POCH S.A. Sigma-Aldrich Sigma-Aldrich

0.99 0.97 0.99

twice distilled twice distilled twice distilled

1.594 1.952 1.491

density density density

Standard uncertainties u are u(ρ) = 0.005 g/cm3, u(T) = 0.3 K and u(p) = 5 kPa. 3094

DOI: 10.1021/acs.jced.5b00131 J. Chem. Eng. Data 2015, 60, 3093−3100

Journal of Chemical & Engineering Data

Article

Figure 3. DSC scan for bulk C6H4Br2/CCl4 mixtures for (a) x2 = 0.1 composition; (b) x2 = 0.2; (c) x2 = 0.8.

Figure 2. Capacity C as a function of temperature T for bulk C6H4Br2(2)/CCl4(1) mixtures for (a) x2 = 0.1; (b) x2 = 0.2; (c) x2 = 0.8. T2 corresponds to the melting process; T1 is the temperature at which the transition into the homogeneous mixture is complete.

Table 2. Melting Temperature T for a Bulk C6H4Br2/CCl4 Mixtures for Different Dibromobenzene Mole Fraction (from 0 to 1)a

at which the transition into the homogeneous liquid state is complete, we can construct the phase diagram (T,x2)−p of the mixture. Investigation of the dielectric relaxation processes versus temperature related to molecular motion in particular phases is discussed below. DSC measurements for bulk phase C6H4Br2/CCl4 mixtures were also performed, using a heating rate in the range 0.5 K/ min to 10 K/min. The results obtained were fully reproduceble. The DSC scans are shown in Figure 3a,b,c. Two DSC peaks are observed, which correspond to the transition temperatures T1 and T2 obtained from the dielectric method. In Figure 3b an irregular DSC peak is observed for concentration x2 ≈ 0.2, showing a very narrow temperature range of the phase transitions, similar to that observed in Figure 2b. In Table 2 the values of temperatures of solidus cuve (T1) and liquidus curve (T2) versus concentration x2 obtained for the bulk C6H4Br2/CCl4 mixtures using dielectric method are presented. In Figure 4 the bulk phase diagram for the C6H4Br2/CCl4 mixture is shown; the diagram is typical for a simple eutectic mixture. The line AEB is the liquidus line, constructed from the temperature T1 at which the system is in the liquid phase in equilibrium with a solid phase. The line CD is a solidus line, obtained from the temperature T2; below this temperature the entire system is in the solid state. At the eutectic point E the solid constituents m-C6H4Br2 and CCl4 are in equilibrium with the liquid mixture of composition x2E ∼ 0.2. The melting points of the two components of the C6H4Br2/CCl4 mixtures are different (C6H4Br2 − 266 K and CCl4 − 250.4K), so the

x2

T2/K

T1/K

0.00 0.10 0.20 0.30 0.40 0.50 0.60 0.70 0.80 1.00

218.0 218.0 218.3 218.8 218.1 218.2 218.0 218.0

252.4 225.0 220.5 229.7 237.8 242.7 247.2 254.1 259.0 266.0

a

T2 corresponds to the beginning of the melting process while T1 is the temperature at which the transition into the homogeneous liquid state is complete. Mole fraction, x2; temperature, T, and pressure p = 0.1 MPa. Standard uncertainties u are u(T) = 0.7 K, u(x2) = 0.01 and u(p) = 5 kPa.

eutectic concentration is shifted toward the component having the lower melting temperature. The capacitance curve C(T) as a function of the temperature was also studied for the C6H4Br2/ CCl4 mixture placed in CPG. Because the sample was introduced as a suspension of CPG particles in bulk mixtures, the signals we observed were from both the bulk and confined system. In Figure 5 a,b we present the melting transition for bulk C6H4Br2 as determined by both DRS and DSC. The results show a sudden increase in C(T) at the melting temperature 266 K (Figure 5a) and an endothermic peak in the DSC scan at 266 K, both related to the melting transition of m-C6H4Br2 (Figure 3095

DOI: 10.1021/acs.jced.5b00131 J. Chem. Eng. Data 2015, 60, 3093−3100

Journal of Chemical & Engineering Data

Article

The analysis of the DSC scan (Figure 5d) shows also the melting transitions at 246 and 266 K, which are attributed to the melting in pores and the bulk m-C6H4Br2, respectively. We observed a decrease of the melting temperature of C6H4Br2 in CPG pores of about 20 K relative to the bulk phase. For C6H4Br2/CCl4 mixtures confined in CPG (Figure 6 a,b,c), T1 corresponds to the temperature at the liquidus line, while T2 is the temperature at the solidus line of the bulk mixture. The changes observed in the C(T) function at the lower temperatures T1p and T2p are attributed to the melting in CPG, and correspond to the liquidus and solidus lines for the C6H4Br2/CCl4 mixture confined in CPG. In Figure 6 panels d and e the DSC scan for the melting of the C6H4Br2/CCl4 mixtures of 0.3 and 0.4 mole fraction confined in CPG is presented. We observe four DSC peaks related with the transition temperatures: T1, T2, T1p, and T2p. These transition temperatures are also observed by the dielectric method. The signals at temperatures T1 and T2 correspond to the melting of the bulk mixture, while the two DSC peaks at T1p and T2p are attributed to the melting in pores. The phase transition temperatures corresponding to the solidus/liquidus lines estimated by the DSC method were consistent with those obtained from the analysis of the C(T) function for this system. The temperatures of solidus line T2p and liquidus line T1p for C 6 H 4 Br 2 /CCl4 mixtures confined in CPG for various concentrations of m-C6H4Br2 obtained on the basis of dielectric measurements are presented in Table 3. The phase diagram for the C6H4Br2/CCl4 mixtures confined in CPG compared with its bulk phase diagram (Figure 3) constructed on the basis of our dielectric and DSC measurements is presented in Figure 7. As seen from Figure 7, the phase diagram of the confined mixture is of the same type as for the bulk mixture, that is, that for a eutectic mixture. The melting point of CCl4 confined in CPG with pores of 7.5 nm diameter is 229 K,14 while that for C6H4Br2 in CPG is 246 K. These values of the melting temperatures in pores suggest that the eutectic composition x2pE in pores should be shifted toward the component having the lower melting temperature, CCl4. Our results show x2pE = 0.15 in the pores, lower than the bulk eutectic composition, x2E = 0.2. It is also seen that the melting point of the confined mixture is lower than that of the bulk mixture (the eutectic temperatures, TE, are about 200 K and 220 K for the confined and bulk mixtures, respectively). The depression of the melting temperature of the confined systems relative to the bulk is typical of systems for which the ratio of the fluid/wall to the fluid/fluid interaction is considerably lower than 1. For liquids having a wettability parameter αw much larger than 1, an increase of the melting temperature is generally observed, as is often the case for liquids adsorbed on nanoporous carbons, due to the high atomic density of the pore walls.4 The melting of C6H5Br/CCl4 mixtures confined in ACF pores of mean diameter 1.4 nm was determined using the dielectric method. The C6H5Br/CCl4 bulk mixture also forms an eutectic system,12 with the eutectic concentration x2E = 0.48 and eutectic temperature TE = 210 K, as shown in Figure 8. The melting temperatures of the two components are similar (bromobenzene 244 K and carbon tetrachloride 250.4 K) and the eutectic concentration x2E is close to 0.5. In Figure 9 the capacity C is shown versus T for C6H5Br in ACF. As seen from Figure 9, the melting temperature of bulk C6H5Br is observed as a sudden increase of C(T) at a temperature 244 K; at 271 K

Figure 4. Solid−liquid phase diagram (T,x2) for C6H4Br/CCl4 bulk mixtures.

Figure 5. (a) Capacitance C versus temperature for bulk C6H4Br2; (b) DSC scan for bulk C6H4Br2; (c) capacitance C versus temperature for C6H4Br2 in CPG; (d) DSC scan for C6H4Br2 in CPG with pores of average diameter 7.5 nm.

5b). In Figure 5c and 5d the capacity curve C(T) and DSC scan for m-C6H4Br2 in CPG of diameter of 7.5 nm are presented. As seen from Figure 5c, the C(T) function exhibits two sharp changes: at 246 K, which is attributed to the melting in pores, and at 266 K resulting from the melting of bulk m-C6H4Br2. 3096

DOI: 10.1021/acs.jced.5b00131 J. Chem. Eng. Data 2015, 60, 3093−3100

Journal of Chemical & Engineering Data

Article

Figure 6. Capacity C as a function of temperature T for C6H4Br2/CCl4 mixtures of different concentrations confined in CPG: (a) x2 = 0.15, (b) x2 = 0.4, and (c) x2 = 0.52; DSC scan for C6H4Br2/CCl4 mixtures confined in CPG of (d) x2 = 0.3 and (e) x2 = 0.4.

Table 3. Melting Temperature T for C6H4Br2/CCl4 Mixtures Confined in CPG for Different Dibromobenzene Mole Fraction (from 0 to 1)a x2

T2p/K

T1p/K

0.00 0.10 0.15 0.20 0.30 0.40 0.52 0.55 0.70 0.85 1.00

201.9 201.0 200.1 201.0 200.0 199.0 199.1 200.1 200.0

229.0 209.0 205.0 207.2 208.0 207.0 207.0 208.1 218.9 238.2 246.0

a

T2 corresponds to the beginning of the melting process while T1 is the temperature at which the transition into the homogeneous liquid state is complete. Mole fraction, x2; temperature, T; and pressure, p = 0.1 MPa. Standard uncertainties u are u(T) = 0.7 K, u(x2) = 0.01 and u(p) = 5 kPa.

Figure 7. Solid−liquid phase diagrams (T,x2) for C6H4Br2(2)/CCl4(1) bulk mixtures (solid circles) and mixtures confined in CPG (open circles).

the next phase transition behavior is observed as an increase in C(T), which can be attributed to the melting in pores. The melting temperature of CCl4 confined in ACF pores of diameter of 1.4 nm is 298 K.15 In Figure 10 panels a, b, and c, the C(T) functions for C6H5Br/CCl4 mixtures placed in ACF of various compositions are presented. At temperatures T1 and T2 in Figure 10, we

observe changes of the C(T) function related to phase transitions of the bulk C6H5Br/CCl4 mixtures, typical for liquidus and solidus lines, respectively. The sudden changes observed for C(T) at the high temperatures T1p and T2p are related to the melting in ACF pores. The values of temperatures of the solidus T2p and liquidus T1p line for various concentrations of C6H5Br for C6H5Br/CCl4 3097

DOI: 10.1021/acs.jced.5b00131 J. Chem. Eng. Data 2015, 60, 3093−3100

Journal of Chemical & Engineering Data

Article

mixtures confined in ACF obtained from dielectric measurements are presented in Table 4. Table 4. Melting Temperature T for C6H5Br/CCl4 Mixtures Confined in ACF for Different Bromobenzene Mole Fraction (from 0 to 1)a

Figure 8. Solid−liquid phase diagrams (T,x2) for C6H5Br/CCl4 bulk mixtures; x2 is the bromobenzene mole fraction. Data are taken from ref 12.

x2

T2p/K

T1p/K

0.00 0.10 0.20 0.30 0.40 0.50 0.55 0.60 0.70 0.80 1.00

247.9 249.9 248.0 248.0 240.2 249.1 248.0 241.0 249.8

298.0 273.0 265.1 263.0 261.2 259.9 260.0 248.0 261.0 270.1 271.0

a

T2 corresponds to the beginning of the melting process while T1 is the temperature at which the transition into the homogeneous liquid state is complete. Mole fraction, x2; temperature, T; and pressure, p = 0.1 MPa. Standard uncertainties u are u(T) = 0.8 K, u(x2) = 0.01 and u(p) = 5 kPa.

The phase diagram for the C6H5Br/CCl4 mixtures confined in ACF, constructed on the basis of dielectric measurements, is compared with the diagram for bulk mixtures (Figure 8), and is presented in Figure 11. As follows from Figure 11 the type of phase diagram for C6H5Br/CCl4 mixtures placed in ACF is also eutectic, similar to that for the bulk mixture. The melting point of CCl4 placed in ACF is higher than the melting temperature of C6H5Br confined in ACF, so we can expect that the eutectic composition x2pE in pores will be shifted toward the component with the lower melting point, that is, toward C6H5Br. Our results show that x2pE = 0.6, which is larger than the bulk eutectic concentration, x2E = 0.48. As shown in Figure 11, the melting temperature of the confined mixture is higher than that of the bulk (the TE values are 245 and 210 K for the confined and bulk mixtures, respectively).

Figure 9. Capacity C versus temperature for C6H5Br in ACF pores of diameter 1.4 nm.

Figure 10. Electric capacity C(T) for C6H5Br/CCl4 mixtures confined in ACF of mean pore diameter 1.4 nm for various compositions: (a) x2 = 0.3, (b) x2 = 0.6, and (c) x2 = 0.7. 3098

DOI: 10.1021/acs.jced.5b00131 J. Chem. Eng. Data 2015, 60, 3093−3100

Journal of Chemical & Engineering Data

Article

For the C6H4Br2/CCl4 mixtures confined in CPG a depression of the melting temperature relative to that of the bulk mixture was observed, whereas for C6H5Br/CCl4 mixtures placed in ACF an elevation of the melting point occurred. In conclusion, we note that the effect of confinement in carbon (ACF) and silica (CPG) pore on the melting temperatures of mixtures can be understood in terms of the wetting parameters of these systems.



AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected]. Funding

We thank the National Center of Science of Poland for financial support under Grant DEC-2013/09/B/ST4/03711. K.E.G. thanks the U.S. National Science Foundation under Grant No. CBET 1160151 for support of this research. Figure 11. Solid−liquid phase diagrams (T,x2) for C6H5Br/CCl4 bulk mixtures (solid circles) and mixtures confined in ACF (open circles).

Notes

The authors declare no competing financial interest.



ACKNOWLEDGMENTS We would like to thank M.Sc. A. Sterczyńska for TEM and SEM images of the porous materials used here, which were obtained in the NanoBioMedical Centre in Poznan.

It was found that for the system C6H4Br2/CCl4 mixtures placed in CPG the melting temperatures are lower than those for the bulk mixture. By contrast, the C6H5Br/CCl4 mixture confined in ACF exhibits an elevation of the melting temperature relative to that of the bulk mixture. This result can be explained on the basis of the value of the wetting parameter, αw, for fluids confined in nanopores. The value of αw eq 1 depends strongly on the atomic density of the solid, and on the interlayer spacing and fluid−wall diameter. For carbon this density is 114 nm−2, but for silica16 it is about 57 nm−2. Also the interlayer spacing Δ is smaller for silica than for carbon.16 As a result, the value of αw for a given fluid interacting with carbon is higher than that the same fluid on silica. Both molecular simulation and experimental studies of freezing/ melting of simple fluids in well characterized pores indicate that large αw values lead to an elevation in the melting temperature, while the reverse is true for small αw, the threshold value of αw being close to unity.4,16−18 Thus, we can expect an increase in the melting temperature for the mixtures placed in ACF pores and a decrease of melting temperature for the mixtures in silica pores, as has been observed experimentally here. An elevation of the melting temperature relative to that for the bulk mixture has also been observed for the azeotropic mixture, CCl4/C6H12, confined in ACF pores.9,10 This system also exhibits a large value of αw.



REFERENCES

(1) Gelb, L.; Radhakrishnan, R.; Gubbins, K. E.; SliwińskaBartkowiak, M. Phase Separation in Confined Systems. Rep. Prog. Phys. 1999, 62, 1573−1659. (2) Alba-Simionesco, C.; Coasne, B.; Dosseh, G.; Dudziak, G.; Gubbins, K. E.; Radhakrishnan, R.; Sliwinska-Bartkowiak, M. Effects of Confinement on Freezing and Melting. J. Phys.: Condens. Matter 2006, 18, R15−R68. (3) Radhakrishnan, R.; Gubbins, K. E.; Sliwinska-Bartkowiak, M. Effect of the fluid-wall interaction on freezing of confined fluids: Toward the development of a global phase diagram. J. Chem. Phys. 2000, 112, 11048−11057. (4) Radhakrishnan, R.; Gubbins, K. E.; Sliwinska-Bartkowiak, M. Global phase diagrams for freezing in porous media. J. Chem. Phys. 2002, 116, 1147−1155. (5) Gubbins, K. E.; Long, Y.; Sliwinska-Bartkowiak, M. Thermodynamics of confined nano- phases. J. Chem. Thermodyn. 2014, 74, 169− 183. (6) Sliwińska-Bartkowiak, M.; Sterczynska, A.; Long, Y.; Gubbins, K. E. Influence of microroughness on the wetting properties of nanoporous silica matrices. Mol. Phys. 2014, 112, 2365−2371. (7) Meyer, R. R.; Sloan, J.; Dunin-Borkowski, R. E.; Kirkland, A. I.; Novotny, M. C.; Bailey, S. R.; Hutchinson, J. L.; Greene, M. L. H. Discrete Atom Imaging of One Dimensional Crystals Formed Within Single-Walled Carbon Nanotubes. Science 2000, 289, 1324−1326. (8) Wilson, M. The Formation of Low-Dimensional Ionic Crystallites in Carbon Nanotubes. J. Chem. Phys. 2002, 116, 3027−3041. (9) Cui, B.; Lin, B.; Rice, S. A. Structure and phase transitions in confined binary colloid mixtures. J. Chem. Phys. 2003, 119, 2386− 2398. (10) Coasne, B.; Czwartos, J.; Gubbins, K. E.; Hung, F. R.; SliwinskaBartkowiak, M. Freezing and Melting in Binary Mixtures Confined in a Nanopore. Mol. Phys. 2004, 102, 2149−2163. (11) Coasne, B.; Czwartos, J.; Sliwińska-Bartkowiak, M.; Gubbins, K. E. Effect of Pressure on the Freezing of Pure Fluids and Mixtures Confined in Nanopores. J. Phys. Chem. B 2009, 113, 13874−13881. (12) Czwartos, J.; Coasne, B.; Sliwińska-Bartkowiak, M.; Gubbins, K. E. Melting of mixtures in silica nanopores. Pure Appl. Chem. 2009, 81, 1953−1959. (13) Chelkowski, A. Dielectric Physics; Elsevier: North-Holland, NY,1980.



CONCLUSIONS We have reported the results of experiments on the melting of eutectic mixtures confined in carbon and silica nanopores. The solid−liquid phase diagram of C6H4Br2/CCl4 mixtures confined in controlled pore glass with an average pore size of H = 7.5 nm, and of C6H5Br/CCl4 mixtures confined in activated carbon fibers having a mean pore size of H = 1.4 nm were constructed using the results of dielectric relaxation spectroscopy and differential scanning calorimetry measurements. We observed that the type of phase diagrams for the confined mixtures were the same as those for the bulk mixtures, namely eutectic. It was also observed that the eutectic concentrations for both systems were shifted toward the component having the lower melting point in the pores, that is toward the component having the smaller wetting parameter, αw. These effects were also observed in previous work for confined azeotropic mixtures.10,11 3099

DOI: 10.1021/acs.jced.5b00131 J. Chem. Eng. Data 2015, 60, 3093−3100

Journal of Chemical & Engineering Data

Article

(14) Sliwinska-Bartkowiak, M.; Gras, J.; Sikorski, R.; Radhakrishnan, R.; Gelb, L.; Gubbins, K. E. Phase Transitions in Pores: Experimental and Simulation Studies of Melting and Freezing. Langmuir 1999, 15, 6060−6069. (15) Kaneko, K.; Watanabe, A.; Iiyama, T.; Radhakrishnan, R.; Gubbins, K. E. A remarkable elevation of freezing temperature of CCl4 in graphitic micropores. J. Phys. Chem. B 1999, 103, 7061−7063. (16) Sliwinska-Bartkowiak, M.; Jazdzewska, M.; Gubbins, K. E.; Huang, L. Melting Behavior of Water in Cylindrical Pores: Carbon Nanotubes and Silica Glasses. Phys. Chem. Chem. Phys. 2008, 10, 4909−4919. (17) Radhakrishnan, R.; Gubbins, K. E.; Watanabe, A.; Kaneko, K. Freezing of simple fluids in microporous activated carbon fibers: Comparison of simulation and experiment. J. Chem. Phys. 1999, 111, 9058−9067. (18) Ayappa, K. G.; Mishra, R. K. Freezing of fluids confined between mica surfaces. J. Phys. Chem. B 2007, 111, 14299−14310.

3100

DOI: 10.1021/acs.jced.5b00131 J. Chem. Eng. Data 2015, 60, 3093−3100